CN111643076A - BECT spike intelligent detection method based on multi-channel electroencephalogram signals - Google Patents
BECT spike intelligent detection method based on multi-channel electroencephalogram signals Download PDFInfo
- Publication number
- CN111643076A CN111643076A CN202010404580.6A CN202010404580A CN111643076A CN 111643076 A CN111643076 A CN 111643076A CN 202010404580 A CN202010404580 A CN 202010404580A CN 111643076 A CN111643076 A CN 111643076A
- Authority
- CN
- China
- Prior art keywords
- spike
- candidate
- channel
- detection
- bect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 75
- 238000007637 random forest analysis Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 22
- 239000013598 vector Substances 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000003044 adaptive effect Effects 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims abstract description 3
- 238000000034 method Methods 0.000 claims description 24
- 238000012360 testing method Methods 0.000 claims description 6
- 230000000630 rising effect Effects 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 230000008030 elimination Effects 0.000 claims 4
- 238000003379 elimination reaction Methods 0.000 claims 4
- 238000003066 decision tree Methods 0.000 description 8
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 210000004556 brain Anatomy 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 206010015037 epilepsy Diseases 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000013138 pruning Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 208000002877 Epileptic Syndromes Diseases 0.000 description 1
- 206010061334 Partial seizures Diseases 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 230000007177 brain activity Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000003710 cerebral cortex Anatomy 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000001037 epileptic effect Effects 0.000 description 1
- 230000004424 eye movement Effects 0.000 description 1
- 201000007186 focal epilepsy Diseases 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4094—Diagnosing or monitoring seizure diseases, e.g. epilepsy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Physiology (AREA)
- General Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Psychiatry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Neurology (AREA)
- Neurosurgery (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides a BECT spike wave intelligent detection method based on a multi-channel electroencephalogram signal, which comprises the following steps: (1) collecting electroencephalogram (EEG), and establishing an experimental database; (2) data preprocessing: performing band-pass filtering on the acquired original EEG data to obtain a standard EEG signal; (3) and (5) performing candidate spike detection, performing adaptive template matching by using the screened class center as a new template, and adding all matching results to obtain a candidate spike. (4) Eliminating false detection spike waves: firstly, two related BP channels of each candidate spike are determined according to the candidate detection result of the AV channel, and then the candidate spikes without the phenomenon of 'needle-tip opposition' on the two related BP channels are removed. (5) Spike wave feature extraction: after eliminating false detection spines, calculating 10 characteristics of each channel; (6) random forest classification: training a random forest classification model by taking the extracted spike wave characteristics as feature vectors; and inputting the spike wave characteristics of the electroencephalogram signals to be analyzed into a random forest model to obtain a BECT spike wave detection result.
Description
Technical Field
The invention relates to the field of computers, in particular to a BECT spike wave intelligent detection method of multi-channel electroencephalogram signals.
Background
Benign epilepsy in children with central temporal zone spike (BECT) is one of the most common epilepsy syndromes in children, the main attack group is school-age children, and the attack number accounts for 15% -20% of children's epilepsy.
Electroencephalograms (EEG) are formed by the discharge of a large number of neurons occurring simultaneously during brain activity, are the most direct reflection of cerebral cortex discharge, and contain a large amount of physiological and disease information. Clinical studies indicate that BECT is primary focal epilepsy, has unknown etiology, is a special form of epilepsy and is mainly manifested by partial discharges in the central and temporal regions. Clinically, BECT patients are diagnosed mainly by detection and quantitative analysis of spikes.
Spike waves are typical characteristic waves in BECT electroencephalogram signals, and are sharp in waveform, high in amplitude and fast in change relative to background waveforms. However, this method is time-consuming, has a high false detection rate, and cannot ensure the accuracy of the detection result, so that the spike automatic detection technology has received more and more attention in recent years.
Although there have been many studies on spike detection methods, more advanced automatic spike detection is still difficult for several reasons. First, the behavior of spikes is different due to individual differences from person to person, and it is difficult to automatically detect spikes in electroencephalogram signals in a simple and consistent manner. Secondly, artifacts caused by factors such as heartbeat, eye movement and muscle movement inevitably exist in the electroencephalogram signals, and the performance of the automatic spike detection method is greatly influenced. Finally, compared with the fully discharged spike, the waveform characteristics of the incompletely discharged spike are not obvious enough, further increasing the difficulty of automatic detection.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a BECT spike intelligent detection method based on multi-channel electroencephalogram signals, so as to improve the identification rate of the BECT spike.
In order to achieve the purpose, the invention is realized by the following scheme: a BECT spike intelligent detection method based on multi-channel electroencephalogram signals comprises the following steps:
step S1: acquiring electroencephalogram signals, selecting an experimental object, establishing a BECT electroencephalogram database, and marking spikes in each channel of the electroencephalogram signals;
step S2: preprocessing the electroencephalogram signals, and removing high-frequency components and artifacts by using a 5-order Butterworth band-pass filter.
Step S3: and (3) performing candidate spike detection by adopting a self-adaptive template matching method.
Step S4: false detection and false detection of spikes.
Step S5: and extracting candidate spike waveform characteristics.
Step S6: and realizing BECT spike detection by adopting a random forest classifier.
According to an embodiment of the present invention, the sampling frequency in step S1 is 500Hz, and a large amount of electroencephalogram data is required to be taken as experimental samples, and the experimental samples include children patients with different sexes and different ages.
According to an embodiment of the invention, in the process of candidate detection for adaptive template matching, morphological characteristics such as rising edge slope, falling edge slope, amplitude and duration of a spike wave marked manually are counted, and a universal template is established.
According to an embodiment of the present invention, in the process of performing candidate spike detection by using an adaptive template matching method, the method includes:
step S31, counting the characteristics of rising edge slope, falling edge slope, amplitude height, duration and the like of the spike waveform in the electroencephalogram data, and defining a universal template;
step S32, setting the window width to 300, and carrying out general template matching operation on the electroencephalogram signals according to the time sequence to obtain candidate spike signals;
step S33, clustering the candidate peaks by adopting a K-means method, and dividing the candidate peaks into different classes according to different waveforms;
step S34, counting the number of candidate spikes in each spike cluster, if the number is less than 5% of the total number of candidate spikes, rejecting the class, and finally taking the centroid of the rest classes as a new template;
and step S35, performing new template matching by respectively using the centroid of each class as a template, and superposing the results to obtain candidate spike detection results.
According to an embodiment of the present invention, in the process of eliminating the false detection spike, firstly, two relevant Bipolar (BP) channels of each candidate spike are determined according to a candidate detection result of an electroencephalogram signal of a unipolar (AV) channel, and then the false detection spike is eliminated according to a "spike-to-spike" phenomenon on the BP channel.
In the process of extracting the candidate spike waveform features, 10 features of the candidate spikes in each channel are calculated, including duration interval, amplitude, slope and area.
According to an embodiment of the present invention, in the process of performing BECT spike detection by using a random forest classification model, the method includes:
step S61: and constructing a feature vector of each candidate spike according to the extracted multi-channel electroencephalogram signals, randomly dividing the feature vector into a training set and a testing set, and training a plurality of decision trees in a random forest classifier by using a plurality of electroencephalogram signal samples in the training set to form a random forest model.
Step S62: and inputting the data in the test set into the trained random forest model to obtain a spike detection result based on a machine learning method.
According to an embodiment of the invention, in the training of the random forest model, the method comprises the following steps:
step S611: the new training set with the same number of samples as the training set is extracted and put back in the training sample set.
Step S612: randomly sampling without playback in the feature vector set to form a feature vector set to be selected;
step S613: and (4) according to the candidate feature training set obtained in the step (S612), calculating the optimal splitting mode of each node and splitting the node without pruning until the impurity degree of each leaf node reaches the specified requirement to form a decision tree.
Step S614: and repeating the steps S611 to S613 until all the decision trees are generated and integrated to obtain the random forest model.
Compared with the prior art, the invention has the following technical effects:
(1) the invention collects the electroencephalogram signal in real time, can detect the spike wave of the patient in time, has simple processing, low cost and wide application prospect, and is easy to popularize;
(2) candidate spike detection is carried out by adopting a self-adaptive template matching method, so that a large amount of false detection spikes can be removed;
(3) eliminating false positive results in the candidate spike waves according to the 'needle point opposite' phenomenon of the candidate spike waves on two related BP channels;
(4) the method extracts the waveform characteristics on each candidate AV channel and two related BP channels to form a multi-channel characteristic vector, thereby ensuring the integrity of spike wave characteristics;
(5) the method extracts the multi-channel electroencephalogram spike wave characteristics from the candidate spike waves, uses the random forest classification model for identification, adds random attributes, and has the advantages of high training speed, strong generalization capability of the model and better classification effect than other classifiers.
Drawings
FIG. 1 is a general flowchart of the BECT spike wave intelligent detection method based on multi-channel electroencephalogram signals.
FIG. 2 is a diagram of an electrode layout according to the present invention.
FIG. 3 is a flowchart illustrating a spike candidate detection process according to the present invention.
FIG. 4 is a diagram illustrating the spike of the AV channel and the BP channel according to the present invention.
FIG. 5 is a schematic diagram of spike feature extraction according to the present invention.
FIG. 6 is a flow chart of the random forest model training of the present invention.
Detailed Description
Electroencephalogram signals usually contain a lot of physiological information about human diseases, and play an important role in the diagnosis and detection of BECT diseases of children. Spike is a typical waveform of BECT, so spike detection is required for brain electrical signals for better research. The existing spike method is difficult to completely and accurately determine the spike position, so that the research on epileptic diseases is greatly influenced. In view of this, the present embodiment provides an intelligent BECT spike detection method based on multi-channel electroencephalogram signals.
In order to make the objects, implementations and innovations of the present invention more prominent, the present invention will be further described in detail with reference to the accompanying drawings and examples.
Fig. 1 is a general flowchart of the intelligent detection method for the BECT spike based on the multi-channel electroencephalogram signal, which includes:
step S1: acquiring an electroencephalogram signal: selecting an experimental object, collecting 19-channel AV lead electroencephalogram data of a BECT patient by using electroencephalogram collection equipment, and establishing an experimental database.
Step S2: data preprocessing: butterworth bandpass filtering the acquired raw EEG data to obtain a standard EEG signal.
Step S3: and (3) candidate spike detection: firstly, defining a universal template according to the waveform characteristics of BECT spike waves, and carrying out universal template matching on 19 AV channel electroencephalogram signals to obtain candidate spike signals; then clustering the candidate peaks by using a K-means algorithm to obtain a plurality of classes; counting the number of candidate peaks in each class, and rejecting the class if the number of spikes is less than 5% of the total number of peaks; and respectively using the screened class centers as new templates to perform adaptive template matching, and adding all matching results to obtain candidate spike detection results.
Step S4: eliminating false detection spike waves: firstly, two related BP channels of each candidate spike are determined according to the candidate detection result of the AV channel electroencephalogram signals, and then false detection spikes are eliminated according to the 'spike-to-spike opposition' phenomenon on the BP channels.
Step S5: spike wave feature extraction: in performing spike feature extraction, 10 features of the candidate spikes in each channel are calculated, including duration interval, amplitude, slope, and area.
Step S6: random forest classification: training a random forest classification model by taking the extracted spike wave characteristics as feature vectors; and inputting the spike wave characteristics of the electroencephalogram signals to be analyzed into a random forest model to obtain a BECT spike wave detection result.
The intelligent BECT spike detection method based on multi-channel electroencephalogram signals provided by the embodiment is described in detail below with reference to FIGS. 1 to 6.
The intelligent BECT spike detection method based on the multichannel electroencephalogram signals provided by the embodiment starts in step S1, wherein a multi-lead electroencephalograph is used for collecting long-range monitoring electroencephalogram signals of a patient, the sampling frequency is 1000Hz, the electrode distribution adopts the international 10-20 electroencephalogram collection standard, 19 channels of electroencephalogram data are collected together, and the electrode distribution is shown in fig. 2. Spike waveforms in each channel of the brain electrical signal are labeled by a professional electroencephalograph.
Then, step S2 is executed to perform preprocessing operation on the brain wave. A5-order Butterworth band-pass filter is adopted to filter frequency components above 50Hz and below 0.5Hz, and the interference of noise and artifacts is reduced.
And step S3, carrying out self-adaptive template matching on the electroencephalogram signals after the preprocessing operation to obtain candidate spike detection results. The candidate spike detection method will be described in detail below with reference to fig. 3.
Firstly, statistical analysis is carried out on spike waveforms marked in the electroencephalogram signal, and the average values of the rising edge slope, the falling edge slope, the peak value and the duration of all marked spike waveforms are respectively obtained and used as standards to establish a universal template (step S31). Then, setting the window width to 300, and performing a general template matching operation on the electroencephalogram signals in time sequence to obtain candidate spikes (step S32). K-means clustering is performed on the candidate peaks, and the candidate peaks are classified into different classes according to the waveform (step S33). And counting the number of candidate peaks in each peak cluster, if the number is less than 5% of the total number of the candidate peaks, rejecting the class, and finally taking the centroid of the remaining class as a new template (step S34). And (4) performing new template matching by using the centroid of each class as a template, and superposing the results to obtain a candidate spike detection result (step S35).
And step S4, eliminating false detection spike waves according to the EEG signals of the AV channel and the two related BP channels. The method of eliminating false detection spikes will be described in detail below with reference to fig. 4 and specific examples.
By analyzing the EEG signal of the AV channel, spikes can be classified into three types. Taking three AV channels F3-AV, C3-AV and P3-AV as examples, if a spike is detected only on C3-AV, the two related BP channels are F3-C3 and C3-P3; if spikes were detected on both C3-AV and P3-AV, the two related BP channels were F3-C3 and P3-O1, respectively; if spikes were detected on F3-AV, C3-AV and P3-AV, the two related BP channels were FP1-F3 and P3-O1. All candidate spikes are checked against this rule and candidate spikes that do not comply with this rule are eliminated. On the BP channel, the spike appears as "opposite spike" as shown in fig. 4.
Step S5 extracts the waveform features of the candidate spike. The candidate spike extraction method will be described in detail below with reference to fig. 5 and a specific example.
Spike waveform characteristics include: duration, amplitude, slope and area. Wherein the duration characteristics include Dur _ left, Dur _ right, and Dur _ spike, which represent spike left half-wave duration, spike right half-wave duration, and spike duration, respectively, corresponding to FIG. 5And
the amplitude characteristics include Amp _ left, Amp _ right, and Amp _ spike, which represent the spike left half-wave amplitude, spike right half-wave amplitude, and spike amplitude, respectively, corresponding to FIG. 5 And
the Slope characteristics include Slope _ left, Slope _ right, and Slope _ spike, which represent the spike left-half Slope, spike right-half Slope, and spike kurtosis, respectively, corresponding to those in FIG. 5And
the Area characteristic is Area _ spike, which represents the spike Area, and the calculation formula isEach candidate spike has 10 feature parameters on one channel, feature parameters on one AV channel and two BP channels of each candidate spike are extracted to form feature vectors, and each feature vector comprises 30 feature parameters.
And step S6, spike wave automatic detection is realized through a random forest classification model. The random forest classification method will be described in detail below with reference to fig. 6.
Step S61 is a training process of the random forest model, and the random forest classifier includes a plurality of decision trees, and its output class is determined by the maximum votes in the results of all the trees. And repeatedly and randomly selecting M samples from the M samples in the original training sample set by using a bootstrap resampling technology to generate a new training sample set, and then generating a random forest by using M individual decision tree classifiers.
The data set is divided into a training set and a testing set, and the random forest model training process is described below with reference to fig. 6:
step S611: firstly, sampling with returning is carried out for M times from all the feature vector sets to form a feature set to be selected, and the number of samples in the feature set to be selected is the same as that of the samples in the original feature vector set.
Step S612: secondly, randomly selecting a certain number of feature vectors from the features to be selected, and selecting the optimal features.
Step S613: and (4) according to the candidate feature training set obtained in the step (S612), calculating the optimal splitting mode of each node and splitting the node without pruning until the impurity degree of each leaf node reaches the specified requirement to form a decision tree.
Step S614: and repeating the step S611 to the step S613 until all the decision trees stop growing, and generating a random forest.
And step S42, inputting the electroencephalogram data in the test set into a random forest model, obtaining a spike wave detection result after voting selection by a decision tree, determining an electroencephalogram segment where a spike wave is located, and further determining an electroencephalogram channel where the spike wave is located and a time point.
The foregoing is considered as illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the scope of the invention in any way, and therefore the present invention is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. A BECT spike intelligent detection method based on multi-channel electroencephalogram signals is characterized by comprising the following steps:
step S1: collecting electroencephalogram signals; selecting an experimental object, acquiring multi-channel electroencephalogram data of a BECT patient by using electroencephalogram acquisition equipment, and establishing an experimental database;
step S2: preprocessing data; performing Butterworth band-pass filtering on the acquired original EEG data to obtain a standard EEG signal;
step S3: detecting candidate spikes; firstly, defining a universal template according to the waveform characteristics of the BECT spike, and carrying out universal template matching to obtain a candidate spike signal; then clustering the candidate peaks by using a K-means algorithm to obtain a plurality of classes; counting the number of candidate peaks in each class, and if the number of peaks is less than 5% of the total number of peaks, rejecting the class; respectively using the screened class centers as new templates to perform adaptive template matching, and adding all matching results to obtain candidate spike detection results;
step S4: eliminating false detection spike waves; firstly, determining two related BP (back propagation) channels of each candidate spike according to a candidate detection result of an AV channel electroencephalogram signal, outputting a time sequence of the candidate spikes on the AV channel, then carrying out spike detection on the two BP channels, judging whether the candidate spikes are detected by the AV channel and the two BP channels at the same time and whether the two candidate spikes on the BP channels are opposite to each other, and if the two candidate spikes are not consistent with the two conditions, rejecting the candidate spikes in a candidate detection result;
step S5: spike wave feature extraction; after FPS elimination, 10 features per channel are calculated for subsequent classification; these 10 features are classified into 4 classes, including duration, amplitude, slope, and area;
step S6: random forest classification: training a random forest classification model by taking the extracted spike wave characteristics as feature vectors; and inputting the spike wave characteristics of the electroencephalogram signals to be analyzed into a random forest model to obtain a BECT spike wave detection result.
2. The method for intelligently detecting BECT spike based on multi-channel EEG signal as claimed in claim 1, wherein in step S2, a 5 th order IIR Butterworth band pass filter with frequency range of 0.5-50Hz is used to remove noise and artifacts in EEG signal during data preprocessing.
3. The intelligent BECT spike detection method based on multi-channel electroencephalogram signals according to claim 1, wherein the step S3 further comprises:
step S31, counting the characteristics of rising edge slope, falling edge slope, amplitude height, duration and the like of the spike waveform in the electroencephalogram data, and defining a universal template;
step S32, setting window width, and carrying out general template matching operation on the electroencephalogram signals according to time sequence to obtain candidate spike signals;
step S33, clustering the candidate peaks by adopting a K-means algorithm, and dividing the candidate peaks into different classes according to different waveforms;
step S34, counting the number of candidate peaks in each peak cluster, if the number is less than 5% of the total number of candidate peaks, rejecting the cluster, and finally taking the centroid of the rest clusters as a new template;
and step S35, performing new template matching by respectively using the centroid of each class as a template, and superposing the results to obtain candidate spike detection results.
4. The intelligent BECT spike detection method based on multi-channel EEG signal as claimed in claim 1, wherein in step S4, when false detection spike elimination is performed, firstly two related BP channels of each candidate spike are determined according to the candidate detection result of the EEG signal of the AV channel, time sequence of the candidate spike on the AV channel is output, then spike detection is performed on the two BP channels, whether the candidate spikes are detected by the AV channel and the two BP channels at the same time or not is judged, whether the two candidate spikes on the BP channels are in "spike opposition" or not is judged, and if the two candidate spikes are not in agreement, the spike elimination is performed in the candidate detection result.
5. The method according to claim 1, wherein in step S5, 10 features of candidate spikes in each channel are calculated during spike feature extraction, including duration interval, amplitude, slope and area.
6. The intelligent BECT spike detection method based on multi-channel electroencephalogram signals as claimed in claim 1, wherein in the step S6, the random forest classification method further comprises:
step S61, constructing a feature vector of each candidate spike according to the extracted spike feature parameters, dividing the feature vector into a training set and a testing set, and training a random forest classification model by using data in the training set;
step S62, inputting the data in the test set into the random forest model, and obtaining an output result, which is a spike detection result, so as to detect whether there is a spike in this segment.
7. The intelligent BECT spike detection method based on multi-channel electroencephalogram signals, as recited in claim 4, wherein said elimination of false spike detection requires comparison with detection results of adjacent channels, if a detection result has a "needle-tip-to-needle" phenomenon in an adjacent channel, the detection result is regarded as a spike, otherwise, the result is discarded.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010404580.6A CN111643076A (en) | 2020-05-13 | 2020-05-13 | BECT spike intelligent detection method based on multi-channel electroencephalogram signals |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010404580.6A CN111643076A (en) | 2020-05-13 | 2020-05-13 | BECT spike intelligent detection method based on multi-channel electroencephalogram signals |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111643076A true CN111643076A (en) | 2020-09-11 |
Family
ID=72342804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010404580.6A Pending CN111643076A (en) | 2020-05-13 | 2020-05-13 | BECT spike intelligent detection method based on multi-channel electroencephalogram signals |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111643076A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113974653A (en) * | 2021-11-30 | 2022-01-28 | 杭州妞诺霄云大数据科技有限公司 | Optimized spike detection method and device based on Joyston index, storage medium and terminal |
CN114027854A (en) * | 2021-12-01 | 2022-02-11 | 杭州电子科技大学 | BECT spike detection method based on optimal template matching and morphological feature extraction |
CN117918861A (en) * | 2024-03-20 | 2024-04-26 | 之江实验室 | Real-time detection system of epileptic single-channel brain electrical signals |
CN117918861B (en) * | 2024-03-20 | 2024-06-04 | 之江实验室 | Real-time detection system of epileptic single-channel brain electrical signals |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110321783A (en) * | 2019-05-07 | 2019-10-11 | 北京慧脑云计算有限公司 | A kind of MEG spike detection method and system based on 1D convolutional neural networks |
CN110477908A (en) * | 2019-07-16 | 2019-11-22 | 华南师范大学 | A kind of method of intelligent recognition epileptogenic focus |
CN110811609A (en) * | 2019-10-28 | 2020-02-21 | 杭州电子科技大学 | Intelligent epileptic spike detection method based on fusion of adaptive template matching and machine learning algorithm |
-
2020
- 2020-05-13 CN CN202010404580.6A patent/CN111643076A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110321783A (en) * | 2019-05-07 | 2019-10-11 | 北京慧脑云计算有限公司 | A kind of MEG spike detection method and system based on 1D convolutional neural networks |
CN110477908A (en) * | 2019-07-16 | 2019-11-22 | 华南师范大学 | A kind of method of intelligent recognition epileptogenic focus |
CN110811609A (en) * | 2019-10-28 | 2020-02-21 | 杭州电子科技大学 | Intelligent epileptic spike detection method based on fusion of adaptive template matching and machine learning algorithm |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113974653A (en) * | 2021-11-30 | 2022-01-28 | 杭州妞诺霄云大数据科技有限公司 | Optimized spike detection method and device based on Joyston index, storage medium and terminal |
CN113974653B (en) * | 2021-11-30 | 2024-03-22 | 杭州妞诺霄云大数据科技有限公司 | Method and device for detecting optimized spike based on about log index, storage medium and terminal |
CN114027854A (en) * | 2021-12-01 | 2022-02-11 | 杭州电子科技大学 | BECT spike detection method based on optimal template matching and morphological feature extraction |
CN117918861A (en) * | 2024-03-20 | 2024-04-26 | 之江实验室 | Real-time detection system of epileptic single-channel brain electrical signals |
CN117918861B (en) * | 2024-03-20 | 2024-06-04 | 之江实验室 | Real-time detection system of epileptic single-channel brain electrical signals |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110811609B (en) | Epileptic spike intelligent detection device based on self-adaptive template matching and machine learning algorithm fusion | |
CN110338786B (en) | Epileptic discharge identification and classification method, system, device and medium | |
CN104783782B (en) | Electrocardiosignal automatic testing method and device | |
CN111772585B (en) | Data extraction method for epileptic abnormal group activities in interval intracranial electroencephalogram signals | |
CN108511055B (en) | Ventricular premature beat recognition system and method based on classifier fusion and diagnosis rules | |
WO2013063053A1 (en) | Method of building classifiers for real-time classification of neurological states | |
CN111000555B (en) | Training data generation method, automatic recognition model modeling method and automatic recognition method for epileptic electroencephalogram signals | |
CN111091074A (en) | Motor imagery electroencephalogram signal classification method based on optimal region common space mode | |
CN114532993A (en) | Automatic detection method for electroencephalogram high-frequency oscillation signals of epileptic | |
CN111643076A (en) | BECT spike intelligent detection method based on multi-channel electroencephalogram signals | |
CN105574348B (en) | A kind of heart beat cycle acquisition methods and device based on BCG signals | |
CN113208634A (en) | Attention detection method and system based on EEG brain waves | |
CN115486855A (en) | Electrocardiogram heart beat classification method based on QRS wave group indefinite cycle superposition | |
CN111067513B (en) | Sleep quality detection key brain area judgment method based on characteristic weight self-learning | |
Mahesh et al. | ECG arrhythmia classification based on logistic model tree | |
CN114587381A (en) | Spike detection method based on multi-channel electroencephalogram intelligent screening and weighted sample generation | |
CN111887811B (en) | Brain abnormal discharge detection method and system based on electroencephalogram signal characteristics | |
CN114376524A (en) | Consciousness assessment method and system based on voice statistic learning | |
Cheng et al. | Multiview feature fusion representation for interictal epileptiform spikes detection | |
CN113974653A (en) | Optimized spike detection method and device based on Joyston index, storage medium and terminal | |
Al-Fahoum et al. | ECG arrhythmia classification using simple reconstructed phase space approach | |
CN113180705A (en) | Fatigue detection method and system based on EEG brain waves | |
Benlamri et al. | An automated system for analysis and interpretation of epileptiform activity in the EEG | |
CN111543977A (en) | Multi-cascade artificial intelligence vagina discharge method based on 12-lead resting electrocardiogram | |
Cho et al. | A New Template Matching Method using Variance Estimation for Spike Sorting. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200911 |